The US and Russia removed a requirement for human review of AI-generated targets during UN negotiations on lethal autonomous weapons this month, according to sources. The two nations stripped multiple safeguards from the landmark accord.
During negotiations at the United Nations this month, American and Russian diplomats worked to weaken a proposed pact regulating lethal autonomous weapons systems, according to sources speaking with the Washington Post.
The most significant change involved eliminating a requirement that humans must review AI-generated targeting decisions before weapons deployment. This safeguard represented a critical control mechanism intended to maintain human oversight in autonomous military systems.
Beyond the targeting review removal, the two nations negotiated to strip away additional protections from the broader agreement. The exact nature of these other safeguards was not detailed, but their elimination suggests a broader pattern of reducing oversight requirements.
The developments mark a setback for international efforts to regulate autonomous weapons technology amid rising concerns about AI militarization. Proponents of strong safeguards have warned that removing human review requirements could enable weapons systems to select and engage targets without human decision-making at critical moments.
The negotiations reflect competing interests among major powers regarding autonomous weapons development. Military applications of AI remain a strategic priority for both the US and Russia, creating tension with international regulatory efforts.
The UN initiative represents an attempt to establish global norms around lethal autonomous weapons before the technology becomes more prevalent. Previous discussions have centered on preventing scenarios where machines independently determine which targets to strike.
The agreement, though weakened, still aims to provide some framework for AI weapons oversight. However, the removal of key safeguards raises questions about the pact's effectiveness in constraining autonomous weapons development and deployment.
Further negotiations on the AI weapons accord are expected as other nations weigh in on the proposed regulations. The outcome could influence how autonomous military systems are developed and used globally in coming years.
Meta unveiled the Muse, a charm-sized device featuring a cute avatar interface capable of handling emails, travel bookings, and shopping tasks. The launch follows backlash over the company's smartglasses and represents a strategic pivot toward a more appealing form factor.
Microsoft has paused rollout of the KB5002907 update after reports that it deactivated or removed perpetual Office 2016 and Office 2019 licenses from user systems.
Matthew Prince, CEO of Cloudflare, is addressing AI's expanding role in shaping internet infrastructure. The company's latest data reveals bots now account for over half of all web traffic, a number that continues to rise.
The University of Oxford has permitted OpenAI to train its AI models on digitized texts from the Bodleian Library, one of the world's oldest and largest academic libraries. Internal documents confirm the historical material has been incorporated into OpenAI's training dataset.